Towards personalized recommendation by two-step modified Apriori data mining algorithm

In this paper a new method towards automatic personalized recommendation based on the behavior of a single user in accordance with all other users in web-based information systems is introduced. The proposal applies a modified version of the well-known Apriori data mining algorithm to the log files of a web site (primarily, an e-commerce or an e-learning site) to help the users to the selection of the best user-tailored links. The paper mainly analyzes the process of discovering association rules in this kind of big repositories and of transforming them into user-adapted recommendations by the two-step modified Apriori technique, which may be described as follows. A first pass of the modified Apriori algorithm verifies the existence of association rules in order to obtain a new repository of transactions that reflect the observed rules. A second pass of the proposed Apriori mechanism aims in discovering the rules that are really inter-associated. This way the behavior of a user is not determined by “what he does” but by “how he does”. Furthermore, an efficient implementation has been performed to obtain results in real-time. As soon as a user closes his session in the web system, all data are recalculated to take the recent interaction into account for the next recommendations. Early results have shown that it is possible to run this model in web sites of medium size.